Introduction

Multiband compression is one of the most versatile tools in a podcast mastering engineer's arsenal. While standard compression affects the entire frequency spectrum equally, multiband compression splits your audio into separate frequency ranges—typically lows, mids, and highs—allowing you to apply independent compression settings to each band. This targeted approach solves specific problems that would otherwise require heavy-handed processing or result in unwanted artifacts. When used correctly, multiband compression can transform a muddy, uneven podcast into a polished, professional broadcast with exceptional clarity and punch. In this guide, we'll walk through everything you need to know: what multiband compression is, when to use it, and exactly how to apply it step by step. Whether you're a seasoned podcaster or just starting to master your own episodes, understanding this technique will elevate your sound. Beyond basic dynamics control, multiband compression gives you surgical precision over your vocal tone, helping your show compete with high-budget productions.

What Is Multiband Compression?

Multiband compression is a dynamic processing technique that splits the audio signal into multiple frequency bands using crossover filters. Each band can be compressed independently with its own threshold, ratio, attack, release, and makeup gain. This is fundamentally different from broadband compression, which applies the same compression to the entire frequency range at once.

For example, a typical three-band setup might divide the spectrum as follows:

  • Low band: 20–150 Hz (handles rumble, bass, and plosive energy)
  • Mid band: 150 Hz–2 kHz (covers most vocal fundamentals and body)
  • High band: 2 kHz–20 kHz (manages sibilance, air, and brightness)

The core advantage is that you can tame problematic frequencies without affecting the rest of the mix. For podcasting, where the voice is the primary element, multiband compression allows you to reduce boominess in the low end while simultaneously adding clarity to the high end—all without making the midrange sound squashed or unnatural. It is particularly effective for speech that varies widely in dynamics or has inconsistent microphone pickup. Some advanced plugins offer four or five bands, but for spoken word three is often the sweet spot between control and complexity.

Note on Crossover Types: Most multiband compressors use Linkwitz-Riley crossover filters (often 12 dB/octave or 24 dB/octave). Steeper slopes provide better band isolation but can introduce phase shifts. For podcast mastering, a 12 dB/octave slope is usually transparent enough—if you hear hollowing or phase artifacts, try increasing the crossover width or switching to a linear-phase mode if your plugin supports it.

When to Use Multiband Compression in Podcast Mastering

Multiband compression is not a silver bullet, but it excels in specific scenarios that are common in podcast audio:

Inconsistent Vocal Dynamics

A single host may speak softly at times and loudly at others. Broadband compression can even out the overall level, but it often over-compresses the louder sections, causing pumping. Multiband compression lets you apply more compression to the frequency ranges that change most dramatically—for instance, the low mids that become boomy when the speaker leans into the mic. By compressing only the low-mid band when it gets loud, the rest of the voice remains natural and breathing room is preserved.

Sibilance and Harshness

Excessive “s” and “sh” sounds can be distracting. A de-esser is essentially a single-band multiband compressor that targets the high frequencies. Using a full multiband compressor gives you more control: you can set a higher ratio in the high band while leaving the rest of the range untouched. You can also narrow the high band to just the sibilant region (around 5–8 kHz) for even more precision, which is especially helpful for voices with aggressive sibilance.

Room Resonance and Muddy Low End

If your recording environment has a low-frequency rumble or boxy resonance, the low band of a multiband compressor can clamp down on that energy without affecting the vocal clarity. This is often more transparent than EQ alone, because the compression reacts dynamically. For example, if your room has a standing wave at 120 Hz, static EQ cut removes that frequency all the time, even when no resonance is present. Multiband compression only attenuates when the resonance triggers the threshold, leaving the natural tone intact during quiet passages.

Punch and Presence

Podcasts that need to cut through on noisy platforms (e.g., in a car or through phone speakers) benefit from a subtle lift in the midrange while the low and high ends are kept under control. Multiband compression can achieve this by using different settings per band. For instance, you might apply a moderate ratio to the mid band to tighten the vocal presence while using a gentler ratio on the lows and highs to keep the overall balance intact.

Uneven Multi-Microphone Recordings

If you’re mixing a podcast with multiple hosts on different mics (e.g., one dynamic, one condenser), multiband compression can help even out the tonal differences. By adjusting each band’s parameters per mic track, you can make voices sound more similar before summing them, reducing the need for heavy EQ later.

Choosing a Multiband Compressor Plugin

Not all multiband compressors are built alike. For podcast mastering, you want a plugin that offers flexible crossover points, low CPU usage, and transparent processing. Popular choices include FabFilter Pro-MB, iZotope Ozone Dynamics, Waves C4, and the free TDR Nova (which works as a dynamic EQ/compressor). Many DAWs also include stock multiband compressors—Logic’s Multipressor, for example, is perfectly capable. When evaluating plugins, listen for:

  • Transparency: Does the compressed version sound natural when bypassed vs. active?
  • Crossover flexibility: Can you adjust both the frequency and the slope?
  • Sidechain options: Some plugins allow external sidechaining, which can be useful for ducking frequencies against another track.
  • Visual feedback: A good real-time spectrum display helps identify where compression is engaging.

For most podcasters, a simple three-band plugin with adjustable crossovers is all you need. Avoid overly complex eight-band solutions that invite over-processing.

Step-by-Step Guide to Applying Multiband Compression

Step 1: Analyze Your Audio

Before touching any plugin, listen critically to your podcast episode. Identify problem areas: Is there a constant low-frequency hum? Do certain phrases sound harsh? Is the voice too thin? Take notes on timestamps. Use a spectrum analyzer to visualize frequency imbalances. Common trouble spots for speech include:

  • 100–250 Hz: Muddy, boxy sound
  • 2–5 kHz: Harshness, sibilance
  • Below 80 Hz: Rumble, mic handling noise

Also note the dynamic range: if the voice varies by more than 10 dB from whispers to shouts, multiband compression can be very beneficial. A loudness meter (like YouLean Loudness Meter) can help quantify the variation.

Step 2: Set Up Your Frequency Bands

Most multiband compressors offer two to four bands. For podcasting, three bands are often sufficient. Start with the default crossover frequencies and adjust based on the specific issues you identified. A good starting point:

  • Band 1 (Low): 20–200 Hz
  • Band 2 (Mid): 200 Hz–2 kHz
  • Band 3 (High): 2 kHz–20 kHz

But feel free to move the crossover points. For example, if the voice is boomy around 150 Hz, set the low-mid crossover at 150 Hz so that band can compress that area independently. If sibilance is narrow (e.g., only at 7 kHz), you might split the high band into two (high-mid and high) for finer control—though this requires a four-band compressor.

Step 3: Adjust Compression Parameters Per Band

For each band, start with gentle settings:

  • Threshold: Set so that compression engages only during peaks or problematic passages. Do not compress entire band constantly. Aim for the reduction meter to move only 2–4 dB on average, with occasional 6 dB reductions on loud peaks.
  • Ratio: 2:1 to 3:1 is a safe starting point for voice. Higher ratios (4:1 or more) can sound unnatural, but are sometimes needed for aggressive sibilance control (use fast attack with higher ratio on the high band only).
  • Makeup gain: Compensate for gain reduction to keep the overall level consistent. Aim for no more than 3–6 dB of gain reduction on any band during loud sections. Use the makeup gain control to bring the band back to unity—listen in solo to ensure the band’s relative level doesn’t shift too much.

Solo each band to hear how it is behaving. Listen for pumping or breathing artifacts, especially in the high band where sibilance can cause rapid gain changes. If you hear pumping, try increasing the release time or lowering the ratio.

Step 4: Fine-Tune Attack and Release

Attack and release are critical for transparency. For speech:

  • Attack: Fast attack times (1–10 ms) catch transient bursts like plosives and sibilance. Slower attacks (20–30 ms) let the initial punch through, which can be useful for preserving vocal presence in the mid band. For the low band, a slow attack (10–20 ms) prevents over-reaction to low-frequency peaks that are often longer in duration.
  • Release: Medium to fast release times (20–100 ms) work well for speech because they allow the compressor to recover between syllables. Too slow and the compression becomes sluggish; too fast and it may cause distortion. For the high band, a release of 20–40 ms helps tame rapid sibilant peaks without creating a "breathing" effect. For the low band, a release of 50–100 ms is often smoother.

Adjust each band individually. Listen to how the compressor responds to a single word: the gain reduction should follow the syllable rhythm, not linger across silence.

Step 5: Listen in Context

Never judge a multiband compressor by soloed bands alone. Listen to the full mix and compare with the original (bypass the plugin). Check for:

  • Naturalness: Does the voice still sound like a person, not a robot?
  • Clarity: Are muddy lows reduced without losing body?
  • Punch: Does the voice cut through without harshness?
  • Consistency: Is the vocal level stable from sentence to sentence?

Make small adjustments—1 dB of threshold change or 5 ms of attack can make a big difference. Use your ears as the ultimate judge. Also test on different playback systems: headphones, laptop speakers, and car audio to ensure the processing translates well.

Advanced Techniques and Creative Uses

De-essing with Multiband Compression

Instead of using a dedicated de-esser, you can set the high band of your multiband compressor to a narrow frequency range (e.g., 5–8 kHz) with a fast attack (1–3 ms) and high ratio (4:1 or higher). This gives you precise control over sibilance without affecting the rest of the high frequencies. Combine with a gentle high-shelf EQ boost (1–2 dB at 10 kHz) if the de-essing makes the voice too dull. Because the compressor only activates on sibilant peaks, the air and brightness remain intact during normal speech.

Adding Punch Without Harshness

To make the podcast sound more energetic, try a subtle upward expansion in the mid band, or use a low ratio (1.5:1) with a fast attack to tighten the sound. At the same time, apply a gentle compression in the high band to prevent harshness from overshooting. The result is a voice that feels more present without being fatiguing. You can also use the low band to tighten the bass—set a ratio of 2:1 with a fast release to keep low-end dynamics consistent without losing the weight.

Controlling Room Resonance

If your recording has a persistent boominess around 120 Hz, set the low band crossover at 100 Hz and 200 Hz. Use a ratio of 3:1 with a threshold that catches only the resonance peaks. This dynamic EQ-like approach is often more natural than static EQ cuts, because it only activates when the resonance is present. It can also help with proxemics changes—when a host moves closer to the mic and the lows become exaggerated, the compressor clamps down exactly when needed.

Multiband Compression for Accent and Dialect Consistency

Some voices have inherent tonal imbalances that vary between words or syllables. For example, a speaker with a strong chest resonance might have inconsistent low-end energy. Multiband compression can smooth these variations by applying more gain reduction in the low band during naturally heavy syllables, creating a more even listening experience.

Common Mistakes to Avoid

  • Over-compressing any single band: More than 6 dB of gain reduction on a band will likely sound unnatural. Use makeup gain carefully to avoid an uneven frequency balance.
  • Setting crossover frequencies based on default presets: Always adjust crossovers to your specific voice and recording environment. A male voice may need a lower low-mid crossover than a female voice.
  • Ignoring phase issues: Some multiband compressors introduce phase shift at crossover points. If something sounds hollow or phasey, try a different plugin or use linear-phase mode if available. Phase issues are most noticeable on headphones.
  • Applying multiband compression before basic cleanup: Remove clicks, pops, and noise floor issues first. Multiband compression will amplify unwanted artifacts, especially in the high band where noise can become more prominent.
  • Relying solely on multiband compression: It works best when combined with EQ, broadband compression, and limiting in a balanced chain. Don't try to solve every problem with one plugin.
  • Setting thresholds too low: If the threshold is so low that compression is active on every syllable, the voice will sound squashed and lifeless. Use the threshold to catch only the peaks that cause problems.

Integrating Multiband Compression with Other Processing

Multiband compression should not be the only processor in your mastering chain. A typical workflow might be:

  1. Noise gate or spectral de-noising to clean the background
  2. EQ to correct tonal balance (e.g., low-cut filter, slight presence boost)
  3. Broadband compression with a gentle ratio (1.5:1 to 2:1) to even out overall dynamics
  4. Multiband compression to target specific frequency problems
  5. Limiter to raise overall loudness without clipping

Always put multiband compression after broadband compression because the broadband compressor already reduces overall dynamic range, making the multiband compressor's job easier and more transparent. Experiment with order, but this sequence works for most spoken-word content. If you find that the broadband compressor is doing too much, consider lowering its ratio and letting the multiband handle the finer issues.

For further reading on compression theory and advanced mastering techniques, check out iZotope's guide to multiband compression, Podcast Engineer's mastering tips, and Sound on Sound's article on multiband compression for speech. These resources offer deeper dives into advanced techniques and troubleshooting. Also consider Universal Audio's explanation of multiband compression for additional perspective on plugin selection.

Conclusion

Multiband compression is an indispensable technique for achieving podcast clarity and punch. By splitting the frequency spectrum into manageable bands, you can solve specific audio issues that broadband compression cannot handle gracefully. Start with gentle settings, listen critically, and adjust one parameter at a time. As you gain experience, you will develop an ear for what each band needs. The key is to aim for transparency—your listeners should hear a cleaner, more engaging podcast, not the processing itself. Practice on a variety of recordings, and soon you'll be able to master any episode with confidence and precision. Your audience will thank you for the professional sound that keeps them coming back for more. Remember that multiband compression is a tool, not a crutch—use it to enhance, not to fix fundamental recording problems. With careful application, it can elevate your podcast to broadcast quality without sacrificing the natural warmth of the human voice.